The ability to predict the length of time from disease onset to major disease outcomes in individual patients with Alzheimer's disease (AD) has implications for patient care, the development of interventions and public health. The major aim of the Predictors Study is to develop prediction algorithms to address this issue. The investigators, who represent three collaborating sites, have collected prospective longitudinal clinical data on two separate cohorts of patients with AD. The Predictors 1 Cohort consists of 236 individuals with AD who have been followed every 6 months for up to 14 years. At this point, 89% of the members of this cohort are deceased. These efforts culminated in the formulation of published algorithms that, for the first time, can reliably estimate the time until an individual patient will require nursing home care or die. Based on data from this cohort, we developed new tools for evaluating clinical features of AD and characterizing disease progression in a second cohort of subjects, the Predictors 2 Cohort. This cohort consists of 264 patients who have been followed for up to 7 years. We propose to continue prospective follow-up of the Predictors 2 cohort to gather the additional information necessary to fully develop predictor models. This additional follow-up will allow us to address the full extent of AD, particularly later stages where significant outcomes such as institutionalization and death are more likely to occur. Our specific aims are to: 1) Refine, extend, and validate our published predictor algorithms by continuing to gather new prospective clinical data; evaluating additional predictors including APOE genotype; neuropsychological profiles, and specific prescribed medications; considering new outcomes, particularly the economic impact of the disease and quality of life; and validating and extend developed algorithms by applying them to the Predictors 2 cohort, and data sets collected at other sites across the country and in Europe; and particularly to data collected from representative, population-based cohorts. 2) Use clinical-pathological studies to examine the relationship of clinical features in the patients to the pathology associated with AD and DLB by quantitating Abeta40 and Abeta42 in different biochemically defined compartments; evaluating high-molecular weight oligomeric forms of alpha-synuclein; and determine the relationship of the quantity and location of these measures with 1) clinical features as well as with 2) quantitative neuropathologic measures that have been collected on an ongoing basis. We will determine whether the accuracy of our prediction algorithms increases when we incorporate clinical features that correlate with these pathologic measures.